JP5605687B2 - Spectral characteristic measuring method, spectral characteristic measuring apparatus, and image forming apparatus having the same - Google Patents

Description

  The present invention relates to a spectral characteristic measuring method for measuring spectral characteristics such as spectral reflectance, spectral transmittance, and parameters (colorimetric values) used in various color systems defined by CIE (International Commission on Illumination). The present invention relates to a characteristic measuring apparatus and an image forming apparatus including the spectral characteristic measuring apparatus.

  Color management such as color stability and color reproducibility is one of the important technical problems in image forming apparatuses such as printing apparatuses and printers. In recent years, image formation is performed using a spectroscope (detection means) such as a spectrophotometer. The color tone management of the apparatus has been performed (for example, Patent Document 1). Specifically, for example, the diffuse reflection light of the output image (measurement target) is measured with a spectroscope to obtain the spectral reflectance, and from there, the XYZ color system, L * a * b * color system, etc. defined by CIE The parameters used in the above are calculated, and the color tone inspection of the output image is performed using this result, and the image forming conditions are adjusted.

FIG. 3 is an explanatory diagram for explaining an example of a spectroscope.
In this measurement system, illumination is performed from the illumination system 2 from a direction inclined by 45 ° from the normal direction of the surface to be measured in the measurement object 1. The spectroscope condenses the diffusely reflected light beam from the surface to be measured on the slit 11 by the condensing lens 3, splits and condenses the light beam that has passed through the slit by the concave diffraction element 12, and guides it to the array light receiving element 13. . The array light receiving element 13 has a plurality of light receiving areas arranged in a direction corresponding to the diffraction direction of the concave diffraction element 12, and detects light intensity signals in different wavelength bands in each light receiving area. Then, the reflectance of each wavelength band corresponding to each light receiving region is calculated from the detection result in each light receiving region of the array light receiving element 13 and the light intensity signal obtained from the standard white surface or the like, and the spectral reflectance is calculated. obtain. If the spectral reflectance obtained in this way is used, other spectral characteristics such as parameters (colorimetric values) such as the XYZ color system and the L * a * b * color system can be grasped.

  The spectroscope that measures visible light, for example, has a light intensity signal discretized into 31 or more wavelength bands detected by spectroscopically detecting a light beam including light in a wavelength range of 400 to 700 [nm] at a pitch of 10 [nm]. The group is output as a sensor response. In such a spectroscope, the diffusely reflected light beam from the surface to be measured is temporally or spatially divided into 31 or more wavelength bands to detect light intensity signals in each wavelength band, and the detected values are sequentially captured. Therefore, it takes a certain amount of time to acquire detection values in all wavelength bands. For this reason, for example, the detection speed is insufficient for an application in which an output image in a recent high-speed image forming apparatus is measured in-line at a speed corresponding to the printing speed, and the application is difficult.

  However, when measuring the spectral characteristics of a measurement object whose spectral reflectance distribution changes relatively gently, such as a printed image, each of the three or more divided sections of about 3 to 16 is used to grasp the spectral characteristics. It is possible to measure the spectral reflectance with a spectroscope that detects the light intensity signal for the wavelength band. For example, a spectral intensity can be obtained by detecting a light intensity signal in a wavelength band of about 3 to 16 with a spectroscope and estimating a detection value in another wavelength band from the detection result. In this case, since the number of light intensity signals to be detected is small, the time required for the detection is short, and the present invention can be applied to applications that require high-speed measurement such as in-line measurement of an output image.

  The spectral reflectance is calculated, for example, by obtaining a conversion matrix from the detected light intensity signal to the spectral reflectance using the measurement results of a plurality of color samples for which the spectral reflectance is obtained in advance. Known conversion matrix calculation methods include a low-dimensional linear approximation method, a Wiener estimation method, an estimation method using a nonlinear operation using a neural network, a multiple regression analysis method, and the like (see Non-Patent Document 1). Non-Patent Document 1 discloses a technique for estimating the spectral reflectance of each point based on a spectral reflectance sample obtained in advance from a multiband image captured using a plurality of broadband color filters. A spectral reflectance estimation method using a multiple regression analysis method with the pixel values of a multiband image and their higher-order terms as explanatory variables and the spectral reflectance as an objective variable is disclosed.

Hereinafter, the calculation method of the transformation matrix will be described by taking the multiple regression analysis method as an example.
Spectral response of the object to be measured is defined as a sensor response vector v that represents a sensor response composed of m light intensity signals in different wavelength bands as a row vector having m elements corresponding to m light intensity signals. If a conversion matrix G is obtained when a row vector composed of l elements representing is a spectral reflectance vector r, the spectral reflectance vector r is estimated from the sensor response vector v using the following equation (1). can do.

Here, a group of spectral reflectance vectors r _{1 to} rn of _n samples (n> m) whose spectral reflectances are known in advance is represented by a sample spectral reflectance matrix R represented by the following equation (2). A group of sensor response vectors v _{1 to} v _n for these samples is represented by a sample sensor response matrix V shown in the following equation (3).

In this case, the regression coefficient matrix (conversion matrix) G of the regression equation from V to R with the sample sensor response matrix V as the explanatory variable and the sample spectral reflectance matrix R as the objective variable is the least square of the sample sensor response matrix V. Using the generalized inverse matrix of Moor-Penrose that gives the minimum norm solution, the following equation (4) can be obtained. Here, the superscript “t” indicates transposition of a vector or a matrix, and the superscript “−1” indicates an inverse matrix. For calculating the inverse matrix, a generally known singular value decomposition method or the like can be used.

なお、特許文献１には、ＲＧＢの３チャンネルのセンサ応答ベクトルを高次項で拡張した例が示されている（特許文献１に記載の式（５）参照）。 In the multiple regression analysis method, when there is a non-linear relationship between the explanatory variable and the objective variable, higher-order terms of the explanatory variable (second and subsequent terms; the same applies hereinafter) are used to improve the estimation accuracy. . Therefore, also in the measurement of the spectral reflectance, it is possible to improve the measurement accuracy by adding a higher-order term of the sensor response. For example, when including second-order term, instead of the sensor response vector v to the m for each wavelength band light intensity signal s ₁ ~s _m only an element, as shown in the following equation (5) An extended sensor response vector v ′ extended by a quadratic term is used.

Note that Patent Document 1 shows an example in which the sensor response vectors of RGB three channels are expanded by high-order terms (see Equation (5) described in Patent Document 1).

  The spectral reflectance calculation method described above measures not only the spectral reflectance, but also other spectral characteristics such as the spectral transmittance of the transmission sample and parameters (colorimetric values) used in the XYZ color system defined by the CIE. In the case of (estimation), the present invention can be similarly applied to a sample in which a sample spectral reflectance matrix R can be created from target variables of a plurality of samples in advance.

However, when the measurement accuracy is improved by using the extended sensor response vector v ′ obtained by extending the sensor response vector v by a high-order term as described above, there are the following problems.
For example, when the number of channels of the spectroscope, that is, the number of elements of the sensor response vector v is 6 (m = 6), the number of secondary terms is 6 × 6 = 36, and thus the extended sensor response vector v ′. The number of elements is 42, which is obtained by adding a first-order term. When the conversion matrix G is calculated for the vector v ′ having such a large number of elements or when the spectral reflectance vector r is calculated, the calculation load is large, and the calculation result is obtained in the required short time. It becomes difficult.

To solve this problem, the present inventors have conducted extensive research results, in the higher order terms (term of the quadratic term later), only contribute little to the estimation of spectral characteristics of such estimation Teisu should spectral reflectance It was also found that there is a term that is likely to reduce the estimation accuracy due to the influence of noise during detection. Therefore, if an extended sensor response vector v ″ that excludes such terms from high-order terms is created and the spectral characteristics are measured using this, the computation time can be reduced compared to the case where this is not excluded. Spectral characteristics can be measured.

  The present invention has been made in view of the above viewpoint, and the object of the present invention is to shorten the calculation time in the case of increasing the measurement accuracy using an extended sensor response including at least a quadratic term of the sensor response, Alternatively, it is to provide a spectral characteristic measuring method, a spectral characteristic measuring apparatus, and an image forming apparatus including the spectral characteristic measuring method capable of further increasing measurement accuracy.

To achieve the above object, a first aspect of the present invention, the light flux incident from the measurement object and split into each predetermined wavelength band, in the spectral characteristic measuring method for measuring the spectral characteristics, each other part of the wavelength band a detection step of detecting at least two light intensity signals are overlapping, the detection using the detected light intensity signal in step, the product of the two light intensity signals some wavelength band overlap each other calculated, and based on the result of the calculation, the extended sensor response generation step of generating an extended sensor response waveband does not contain the product of the two light intensity signals do not overlap, generated by the extended sensor response generation process it is characterized in that it has a spectral characteristic deriving step of deriving the spectral characteristics of the measurement target on the basis of the extended sensor responses.
The invention of claim 2 is the spectral characteristic measuring method according to claim 1, by dispersing the respective light beams incident from the plurality of samples having known spectral characteristics respectively for each of the predetermined wavelength band, each wavelength band a sample detection step of detecting a light intensity signal, by using a light intensity signal detected by the sample detecting step, and a conversion parameter calculating step of calculating the conversion parameters, the spectral characteristic deriving step, the conversion using the transformation parameters parameter calculating process is calculated, and is characterized in that for deriving the spectral characteristics of the measurement object from the extension sensor responses generated by the extended sensor response generation process.
Further, the invention of claim 3 is the spectral characteristic measurement method of claim 1 or 2, wherein the extended sensor response generation step calculates a product of two light intensity signals between wavelength bands in which the center wavelengths are adjacent to each other. , the result of the operation in conjunction with the sensor response, characterized in that it is intended to produce an extended sensor response that does not contain the product of the two light intensity signals between a wavelength band in which the center wavelength not adjacent is there.
The invention of claim 4 is the spectral characteristic measuring method according to any one of claims 1 to 3, wherein the extension sensor response generation step, the two in place of the product of the two light intensity signals using the square root of the product of the light intensity signal, it is characterized in that to calculate the extended sensor response.
The invention of claim 5 is the light flux incident from the measurement object and split into each predetermined wavelength band, in the spectral characteristic measuring apparatus for measuring the spectral characteristics, at least part of the wavelength band overlap each other 2 one of the detecting means for detecting a light intensity signal, by using a light intensity signal detected by the detecting means, calculates the product of the two light intensity signals some wavelength band overlap each other, the operation Based on the result , an extended sensor response generating means for generating an extended sensor response not including the product of two light intensity signals whose wavelength bands do not overlap, and an extended sensor response generated by the extended sensor response generating means it is characterized in that it has a spectral characteristic deriving means for deriving the spectral characteristics of the measurement target based.
Also, the invention of claim 6 is the image forming apparatus for forming an image composed of a plurality of colors on an image bearing medium, the spectral characteristics measuring means for measuring the spectral characteristics of the image formed on the image carrying medium , characterized and an image forming condition adjusting means for adjusting the image forming conditions based on the spectral characteristics measured by spectroscopic characteristic measurement means, as the spectral characteristic measuring means, the use of spectral characteristic measuring device according to claim 5 It is what.

  As a result of diligent research, the present inventors have found that the product of two light intensity signals in which a part of the wavelength band overlaps each other has little or no signal component that can be interacted with. It was found that the inclusion of the response hardly contributes to the improvement of the measurement accuracy of the spectral characteristics, and that the measurement accuracy is highly likely to be lowered due to the influence of noise at the time of detection. Based on this knowledge, the present invention improves the measurement accuracy of spectral characteristics using an extended sensor response including the product (second order term) of two light intensity signals. The product of two light intensity signals that partially overlap each other is included, but the product of two light intensity signals that do not overlap the wavelength band is not included. As a result, the number of elements of the extended sensor response is reduced as compared with the conventional measurement method in which the latter product is included in the extended sensor response and the spectral characteristic of the measurement target is measured, so that the calculation time can be shortened. In addition, the latter product, which is likely to reduce the measurement accuracy, is not included in the extended sensor response, so that the measurement accuracy can be increased.

  As described above, according to the present invention, it is possible to shorten the calculation time when increasing the measurement accuracy using the extended sensor response including at least the second order term of the sensor response, or to improve the measurement accuracy. An effect is obtained.

It is a functional block diagram of the spectral characteristic measuring apparatus in the embodiment. It is a graph which shows six light intensity signals detected with the spectroscope of the same spectral characteristic measuring apparatus. It is explanatory drawing for demonstrating an example of the spectrometer. It is explanatory drawing which shows an example of the image forming apparatus carrying the same spectral characteristic measuring apparatus.

Hereinafter, an embodiment of a spectral characteristic measuring method and a spectral characteristic measuring apparatus according to the present invention will be described with reference to the drawings.
FIG. 1 is a functional block diagram of the spectral characteristic measuring apparatus.
In this functional block diagram, a block portion connected by a solid line arrow is a function used when calculating the spectral characteristic of the measurement object based on the sensor response from the spectroscope 101. On the other hand, the block portion connected by the broken-line arrow is a function used when calculating a conversion matrix which is a conversion parameter used for calculation of spectral characteristics.

First, the configuration and operation of the spectral characteristic measuring apparatus will be described along the flow of measuring the spectral reflectance (spectral characteristic) of the measurement object by the spectral characteristic measuring method using the spectral characteristic measuring apparatus.
The spectroscope 101 includes light receiving regions (light intensity sensors) each corresponding to a light intensity signal of each wavelength band obtained by dispersing a light beam incident from a measurement object into m wavelength bands (wavelength bands). ) And outputs these light intensity signals as sensor responses. In the present embodiment, the spectrometer shown in FIG. The sensor response output from the spectroscope 101 is defined as a sensor response vector v with a row vector having light intensity signals in m wavelength bands as elements. The sensor response output from the spectroscope 101 is input to a sensor response extension unit 102 as an extended sensor response generation unit.

  The sensor response extension unit 102 uses the sensor response input from the spectroscope 101 to calculate a product (second-order specific term) of two light intensity signals in which parts of the wavelength band overlap each other, An extended sensor response is generated by combining the calculation result with the sensor response. At this time, the sensor response extension unit 102 does not calculate the product between the light intensity signals whose wavelength bands do not overlap, and does not include this in the extended sensor response. A row vector whose elements are the light intensity signals of m wavelength bands, which are the elements of the sensor response, and the second-order specific term calculated here is defined as an extended sensor response vector v ″. The extended sensor response generated by the sensor response extension unit 102 in this way is input to the spectral reflectance estimation unit 103 as a spectral characteristic deriving unit.

  The spectral reflectance estimation unit 103 includes an extended sensor response vector v ″ based on the extended sensor response generated by the sensor response extension unit 102, and a conversion matrix G stored in the conversion matrix storage unit 104 serving as a conversion parameter storage unit. ”Is calculated, and the calculation result is output as a spectral reflectance vector r indicating the spectral reflectance of the measurement target. In the present embodiment, this spectral reflectance vector r is derived as the spectral characteristic of the measurement target, and this is used as the measurement result.

  In the present embodiment, regarding the product (second order term) of the first order term that is an element of the extended sensor response vector v ″, when the wavelength bands of the two light intensity signals used for calculation of the second order term do not overlap each other, The two light intensity signals are optically independent and do not interact. On the other hand, when there are overlapping portions in the wavelength bands of the two light intensity signals used for the calculation of the second-order term, the product has an interaction and becomes significant. In the present embodiment, since only the quadratic term that can have such a significant interaction is included in the extended sensor response vector v ″, the measurement accuracy (estimation accuracy) of the spectral reflectance (spectral reflectance vector r) is increased. Will improve.

Next, an example of a method for calculating the transformation matrix G stored in the transformation matrix storage unit 104 will be described.
In the present embodiment, by preparing a plurality of color samples having a known spectral reflectance, a sample thereof the spectral reflectance vector r ₁ ~r _n as a sample spectral reflectance matrix R shown in equation (2) below This is stored in the spectral reflectance storage unit 105. Then, a sensor response (sample sensor response) is obtained by the spectroscope 101 for these color samples.

Thereafter, the sample sensor response of each color sample is input to the sensor response extension unit 102, and the sensor response extension unit 102 extends an extended sensor response vector (sample extended sensor response vector) v ₁ ″ to v _n ′ corresponding to each sample sensor response. Generate '. The group of the sample extended sensor response vectors v _{1 to} v _n is a sample extended sensor response matrix V ″ shown in the following equation (3 ′).

The sample extended sensor response matrix V ″ composed of the sample extended sensor response vectors v ₁ ″ to v _n ″ generated in this way is input to the conversion matrix calculation unit 106 as conversion parameter calculation means. Then, the conversion matrix calculation unit 106 uses the sample extended sensor response matrix V ″ and the sample spectral reflectance matrix R stored in the sample spectral reflectance storage unit 105 to obtain the following equation (4 ′) Thus, a conversion matrix G ″ is calculated. The transformation matrix G ″ calculated in this way is set in the transformation matrix storage unit 104.

FIG. 2 is a graph showing six light intensity signals detected by the spectroscope 101 shown in FIG.
The spectroscope 101 detects the light intensity signal for each wavelength band with the array light receiving element 13 after spectrally separating with the diffraction element 12. In this spectroscope 101, since the width of the slit 11 is finite, the image on the array light receiving element 13 when an arbitrary monochromatic light is input depends on the width of the slit 11 and the imaging performance of the optical system. With a width of Therefore, when the center of gravity of this image is located at the boundary between two light receiving areas in the array light receiving element 13, the light intensity signal is detected in these two light receiving areas. That is, as shown in FIG. 2, the light intensity signals detected in at least two adjacent light receiving regions are partially overlapped with each other in the wavelength band.

In the example shown in FIG. 2, for the combination of two light intensity signals detected in adjacent light receiving areas, the overlapping of the wavelength bands is sufficient, that is, the signal component of the light intensity signal for the overlapping wavelength band part is It is big enough. The overlapping state of the wavelength bands between the adjacent light receiving regions can be adjusted as appropriate by changing the width of the slit 11. On the other hand, there is no insufficient or overlap overlapping wavelength band for other combinations, for the signal component of the light intensity signal of the wavelength band portion adapted heavy is or not too small, the noise at the time of measurement and the signal component Is difficult to distinguish. Therefore, the inclusion of the extended sensor response vector v ″ in the spectral characteristics measurement accuracy (estimation accuracy) in each wavelength band can contribute to the improvement of the spectral characteristics of the second order term consisting of the product of two light intensity signals. band is only the second order term comprising the product of the two light intensity signals overlap sufficiently. In the example shown in FIG. 2, such a significant second-order term is a product of two light intensity signals detected in the light receiving regions adjacent to each other (that is, the wavelength bands where the center wavelengths are adjacent). . Therefore, the extended sensor response vector v ″ in this case is as shown in the following equation (6).

  For example, when the array light receiving element 13 has six (m = 6) light receiving regions, according to the extended sensor response vector v ″ of the present embodiment, the conventional extended sensor response vector v shown in the above equation (5). Only five of the 36 quadratic terms included in 'are included. That is, the number of elements of the extended sensor response vector v ″ in this embodiment is 11, which is much smaller than the number of elements (42) of the conventional extended sensor response vector v ′ shown in the above formula (5). Can do.

[Modification]
Next, a modified example in which the extended sensor response vector v ″ of the above embodiment is changed will be described.
The basic configuration of this modification is the same as that of the above embodiment, but the product of two light intensity signals having wavelength bands adjacent to each other is used as an expansion element used for sensor response expansion in the sensor response expansion unit 102. Instead, the second embodiment is different from the above embodiment in that the square root of the product of two light intensity signals having wavelength bands adjacent to each other is used. Accordingly, the extended sensor response vector v ′ ″ in this modification is as shown in the following equation (6 ′).

When the product of two light intensity signals as in the above embodiment is used as an extension element of the extension sensor response, the product corresponds to a signal related to a signal component with overlapping wavelength bands, but the signal is The signal is approximately proportional to the square of each light intensity signal. However, the spectral reflectance vector r ₁ ~r _n sample spectral reflectance matrix R of interest variables in calculating a transformation matrix G also samples the sensor response vector v ₁ to v sample sensor response matrix V as explanatory variables _{Since n is} also a signal proportional to the light intensity, it is desirable that the signal used as the expansion element is also a signal substantially proportional to the light intensity. From this knowledge, in this modification, the sensor response was expanded using the square root of the product of the two light intensity signals as the expansion element, so that the signals used as the expansion element are in portions where the wavelength bands overlap each other. It has an effect similar to a signal having a central wavelength and approximately proportional to the light intensity. Therefore, the measurement accuracy (estimation accuracy) of the spectral reflectance vector r is improved as compared with the case of the above embodiment.

  In the above description, the spectral reflectance of the sample (sample spectral reflectance matrix R) is used as the objective variable when obtaining the conversion matrix G. Instead, for example, the XYZ color system defined by the CIE is used. Tristimulus values may be used. Even in this case, the tristimulus values of the color samples are separately measured in advance with a spectrocolorimeter or the like, and a matrix similar to the sample spectral reflectance matrix R is created, so that the same applies. This is advantageous in that tristimulus values can be directly measured (estimated) with high accuracy from the sensor response of the spectroscope 101. Similarly, regarding other spectral characteristics such as spectral transmittance, the present invention can be similarly applied by preparing a plurality of samples from which an objective variable can be acquired in advance.

  In the present embodiment (including the above-described modification), the case where the spectroscope 101 using the diffraction element 12 and the array light receiving element 13 is used has been described as an example. However, the present invention is not limited to this. is not. If the spectral characteristics of the output signal from the spectroscope are broadband bandpass filters with different center wavelengths, and there is a part where the center wavelength overlaps part of the wavelength band of adjacent signals, it can be used in the same way it can. For example, a spectroscope that detects light while sequentially replacing a plurality of broadband color filters (for example, a spectroscope as described in FIG. 1 of Non-Patent Document 1) or a plurality of light sources (such as LEDs) having different center wavelengths. A spectroscope that detects light while sequentially lighting can be used. Further, for example, even an RGB color sensor using a color filter is applicable if there is a region overlapping the spectral characteristics of the color filter.

  In addition, the two light intensity signals in which a part of the wavelength bands used as the extension elements overlap each other are not limited to signals whose center wavelengths are adjacent to each other. As long as the light intensity signals partially overlap each other. For example, when each light intensity signal obtained by spectroscopy has two or more wavelength bands separated from each other, that is, when the dispersed light corresponding to each light intensity signal has a peak at two or more wavelength portions, respectively. However, it is also possible to use two light intensity signals whose wavelength bands near the peak wavelength overlap each other.

Further, the spectral characteristic measuring apparatus described above is mounted on an image forming apparatus, and there is a utilization method such as adjusting image forming conditions based on the spectral characteristics of an output image measured by the spectral characteristic measuring means. Hereinafter, this example will be described with reference to FIG.
As shown in FIG. 4, the image forming apparatus 80 includes the above-described spectral characteristic measuring apparatus indicated by reference numeral 60. In addition, the image forming apparatus 80 includes a paper feed cassette 81a, a paper feed cassette 81b, a paper feed roller 82, a controller 83, a writing optical system 84, a photosensitive member 85, an intermediate transfer member 86, a fixing roller 87, and a paper discharge roller 88. And so on. Reference numeral 90 denotes an image bearing medium (paper or the like).

  In the image forming apparatus 80, when the four photoconductors 85 are exposed by the writing optical system 84 based on the image data, an electrostatic latent image corresponding to each color is formed on each photoconductor 85. Development processing is performed by attaching a color material such as toner of a color corresponding to the electrostatic latent image. Each color toner image formed on each photoconductor 85 by this development processing is transferred onto the intermediate transfer body 86 so as to overlap each other. Thereafter, the toner image on the intermediate transfer body 86 is transferred from the paper feed cassettes 81 a and 81 b to the image carrier medium 90 conveyed by a guide (not shown) and the paper feed roller 82. The toner image transferred onto the image bearing medium 90 in this manner is fixed by the fixing roller 87, and the image bearing medium 90 is ejected by the ejection roller 88. In the present embodiment, the spectral characteristic measuring device 60 is installed at the subsequent stage of the fixing roller 87.

  In this image forming apparatus, the spectral characteristic measurement device 60 measures (estimates) the spectral characteristics (spectral reflectance, etc.) of the output image (measurement target) at a predetermined timing, and sends the measurement results to the controller 83. The controller 83 functions as an image forming condition adjusting unit, and adjusts the image forming condition based on the measurement result. As a result, since automatic color calibration is possible, it is possible to continuously provide a high-quality image with little color variation, and to stably operate the image forming apparatus.

As described above, in the spectral characteristic measurement method according to the present embodiment, the light beam incident from the measurement target is dispersed for each predetermined wavelength band, and the light intensity in each wavelength band is obtained by the array light receiving element 13 that is a light intensity sensor. detecting a signal, and measures the spectral reflectance is a spectral characteristic of the measurement object based on this the sensor response consisting of the light intensity signal s ₁ ~s _m obtained (sensor response vector v). In this spectral characteristic measurement method, the array light receiving element 13 detects the light intensity signals s ₁ to s ₁ of each wavelength band so that the array light receiving element 13 detects at least two light intensity signals whose wavelength bands partially overlap each other. a detecting step of spectroscope 101 for detecting the s _m, using the detection step light intensity signal s ₁ ~s _m detected in, of the two light intensity signals some wavelength band overlap each other The product (s ₁ × s ₂ , s ₂ × s ₃ ,..., S _m-1 × s _m ) is calculated, and the calculation result is combined with the sensor response vector v so that the wavelength bands do not overlap. An extended sensor response generation step in the sensor response extension unit 102 that generates an extended sensor response vector v ″ that does not include the product of two light intensity signals, and an extended sensor response vector v ′ generated in the extended sensor response generation step Spectral reflectance of the measurement object based on A spectral characteristic deriving step in the spectral reflectance estimating unit 103 for deriving the reflectance vector r). According to this method, since the number of secondary terms that are the expansion elements constituting the extended sensor response vector is small, it is possible to shorten the calculation time when the measurement accuracy is increased using the extended sensor response. In addition, since the quadratic term used as the expansion element is only a significant term, an effect of increasing the measurement accuracy can be obtained.
Further, in the present embodiment (including the modified example), the respective light beams incident from the plurality of samples each having a known spectral reflectance r ₁ ~r _n spectrally for each of the predetermined wavelength band array receiving A sample detection step of detecting a light intensity signal in each wavelength band by the element 13, and the extended sensor response for each sample is generated using the light intensity signal detected in the sample detection step, and a plurality of extensions generated from the sample extension sensor response matrix V '' and the sample spectral reflectance matrix R formed of the above known spectral reflectance r ₁ ~r _n in each sample of the sensor response, expansion sensors produced by the extended sensor response generating step A conversion parameter calculation step of calculating a conversion matrix G ″ that is a conversion parameter for converting the response vector v ″ to the spectral reflectance vector r to be measured. In the spectral characteristic deriving step, using the conversion matrix G ″ calculated in the conversion parameter calculating step, the spectral of the measurement target is calculated from the extended sensor response vector v ″ generated in the extended sensor response generating step. A reflectance vector r is derived. Thereby, the calibration work of a spectroscope can be performed appropriately.
In this embodiment, in the extended sensor response generation step, the product of two light intensity signals between wavelength bands adjacent to each other in the center wavelength is calculated, and the calculation result is combined with the sensor response vector v to obtain the center An extended sensor response vector v ″ that does not include the product of two light intensity signals between the wavelength bands where the wavelengths are not adjacent to each other is generated. This method is known to be a band-pass filter having different optical characteristics of the sensor response, such as a spectroscope 101 that spectrally separates with a prism or diffraction element 12 and obtains a sensor response with the array light receiving element 13. The sensor response can be expanded by the interaction of the adjacent output signals of the center wavelengths, and the estimation accuracy can be stably increased.
In the modified example, in the extended sensor response generation step, the extended sensor response vector v ′ ″ is calculated using the square root of the product of the two light intensity signals instead of the product of the two light intensity signals. . Thereby, the spectral reflection can be estimated with higher accuracy.

  The spectral reflectance measurement method described above can be realized as software in a computer, for example.

DESCRIPTION OF SYMBOLS 1 Measurement object 2 Illumination system 3 Condensing lens 11 Slit 12 Diffraction element 13 Array light receiving element 60 Spectral characteristic measurement apparatus 80 Image forming apparatus 85 Photoconductor 86 Intermediate transfer body 90 Image carrier medium 101 Spectroscope 102 Sensor response expansion part 103 Spectral reflection Rate estimation unit 104 Conversion matrix storage unit 105 Sample spectral reflectance storage unit 106 Conversion matrix calculation unit

JP 2007-208708 A

Tokudo Tsumura, Hideaki Haneishi, Yoichi Miyake, “Estimation of Spectral Reflectance from Multiband Images by Multiple Regression Analysis”, Optics, Japan Optical Society, 1998, 27, 7, 384-391

Claims (6)

The light flux incident from the measurement object and split into each predetermined wavelength band, in the spectral characteristic measuring method for measuring the spectral characteristics,
A detection step of detecting at least two light intensity signal that part of the wavelength band overlap each other,
Using detection light intensity signal detected in step, it calculates the product of the two light intensity signals some wavelength band overlap each other, based on the result of the calculation, has a wavelength band overlap An extended sensor response generating step for generating an extended sensor response that does not include a product of no two light intensity signals;
A spectral characteristic measuring method comprising: a spectral characteristic deriving step of deriving the spectral characteristic of the measurement object based on the extended sensor response generated in the extended sensor response generating step. In the spectral characteristic measuring method of Claim 1,
The respective light beams incident from the plurality of samples having known spectral characteristics each split into each of the predetermined wavelength band, and the sample detection step of detecting a light intensity signal for each wavelength band,
With a light intensity signal detected by the sample detecting step, and a conversion parameter calculating step of calculating a conversion parameter,
The spectral characteristic deriving step, spectroscopic said transformation parameter calculating process by using the conversion parameter calculated, characterized by deriving the spectral characteristics of the measurement object from the extension sensor responses generated by the extended sensor response generation process Characteristic measurement method. The spectroscopic characteristic measurement method according to claim 1 or 2, wherein the extended sensor response generation step calculates the product of the two light intensity signals between wavelength band centered wavelengths are adjacent, sensor response results of the calculation A spectral characteristic measurement method characterized by generating an extended sensor response that does not include a product of two light intensity signals between wavelength bands in which the center wavelengths are not adjacent to each other. The spectroscopic characteristic measurement method according to any one of claims 1 to 3, wherein the extension sensor response generation step, using the square root of the product of the two light intensity signals in place of the product of the two light intensity signals Te, spectroscopic characteristic measurement method characterized by calculating the extended sensor response. The light flux incident from the measurement object and split into each predetermined wavelength band, in the spectral characteristic measuring apparatus for measuring the spectral characteristics,
A detecting means for detecting at least two light intensity signal that part of the wavelength band overlap each other,
With a light intensity signal detected by the detecting means, calculates the product of the two light intensity signals some wavelength band overlap each other, based on the result of the calculation, has a wavelength band overlap Extended sensor response generating means for generating an extended sensor response that does not include a product of no two light intensity signals;
A spectral characteristic measuring device comprising spectral characteristic deriving means for deriving the spectral characteristic of the measurement object based on the extended sensor response generated by the extended sensor response generating means. In an image forming apparatus for forming an image composed of a plurality of colors on an image bearing medium,
A spectral characteristic measuring means for measuring the spectral characteristics of the image formed on the image carrying medium,
Image forming condition adjusting means for adjusting image forming conditions based on the spectral characteristics measured by the spectral characteristic measuring means;
As the spectral characteristic measuring means, an image forming apparatus which comprises using a spectroscopic characteristic measurement apparatus according to claim 5.

Images